Closed mesunhlf closed 4 years ago
Found this in “tensorflow/contrib/keras/python/keras”
def categorical_crossentropy(output, target, from_logits=False):
Categorical crossentropy between an output tensor and a target tensor.
Arguments:
output: A tensor resulting from a softmax
(unless `from_logits` is True, in which
case `output` is expected to be the logits).
target: A tensor of the same shape as `output`.
from_logits: Boolean, whether `output` is the
result of a softmax, or is a tensor of logits.
Returns:
Output tensor.
when i run the command "python train.py models/modelA --type=0 --num_epochs=6" i meet the error:
Traceback (most recent call last): File "/usr/lib/pycharm-community/helpers/pydev/pydevd.py", line 1596, in
globals = debugger.run(setup['file'], None, None, is_module)
File "/usr/lib/pycharm-community/helpers/pydev/pydevd.py", line 974, in run
pydev_imports.execfile(file, globals, locals) # execute the script
File "/home/memoonhlf/github/blackbox-attacks/train.py", line 61, in
main("models/modelB", 1)
File "/home/memoonhlf/github/blackbox-attacks/train.py", line 41, in main
tf_train(x, y, model, X_train, Y_train, data_gen, None, None)
File "/home/memoonhlf/github/blackbox-attacks/tf_utils.py", line 86, in tftrain
optimizer = tf.train.AdamOptimizer().minimize(out)
File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/training/optimizer.py", line 350, in minimize
([str(v) for , v in grads_and_vars], loss))
ValueError: No gradients provided for any variable, check your graph for ops that do not support gradients, between variables ["<tf.Variable 'convolution2d_1_W:0' shape=(8, 8, 1, 64) dtype=float32_ref>", "<tf.Variable 'convolution2d_1_b:0' shape=(64,) dtype=float32_ref>", "<tf.Variable 'convolution2d_2_W:0' shape=(6, 6, 64, 128) dtype=float32_ref>", "<tf.Variable 'convolution2d_2_b:0' shape=(128,) dtype=float32_ref>", "<tf.Variable 'convolution2d_3_W:0' shape=(5, 5, 128, 128) dtype=float32_ref>", "<tf.Variable 'convolution2d_3_b:0' shape=(128,) dtype=float32_ref>", "<tf.Variable 'dense_1_W:0' shape=(128, 10) dtype=float32_ref>", "<tf.Variable 'dense_1_b:0' shape=(10,) dtype=float32_ref>"] and loss Tensor("Reshape_13:0", shape=(?,), dtype=float32, device=/device:GPU:0).